Modeling Mood Variation Associated with Smoking Using a Mixed-Effects Location Scale Model for Analysis of Ecological Momentary Assessment (EMA) Data

• Activity Number: 219
• Type: Invited
• Date/Time: Monday, August 2, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Society for Research on Nicotine and Tobacco
• Author(s): Donald Hedeker*+ and Robin Mermelstein
• Companies: University of Illinois at Chicago and University of Illinois at Chicago
• Keywords: EMA ; mixed model ; heterogeneous variance ; location-scale ; random scale

Abstract

An important issue in smoking research is the effect of smoking on mood levels/variation as a person progresses in their smoking "career." We present data from an adolescent study in which subjects carry a palm pilot for a week and enter random prompts and event-triggered prompts (when they smoke). Using this Ecological Momentary Assessment data (EMA), we describe how mixed models, including heterogeneous between- and within-subjects variance including random subject scale effects, can be used to address key smoking research questions. For example, this intensive measurement allows us to model differences in both variances, which permits us to investigate the degree to which changes in smoking across time is associated with changes in mood variation attributable to smoking. Also, by including the random scale effect we estimate the heterogeneity in the within-subjects variance.